作者: Kevin B. Korb , Lucas R. Hope , Ann E. Nicholson , Karl Axnick
DOI: 10.1007/978-3-540-28633-2_35
关键词: Causal model 、 Psychological intervention 、 Causal reasoning 、 Cognitive psychology 、 Bayesian network 、 Generalization 、 Machine learning 、 PEARL (programming language) 、 Computer science 、 Causal system 、 Artificial intelligence 、 Causality
摘要: The use of Bayesian networks for modeling causal systems has achieved widespread recognition with Judea Pearl's Causality (2000). There, Pearl developed a "do-calculus" reasoning about the effects deterministic interventions on system. Here we discuss some different kinds intervention that arise when indeterminstic are allowed, generalizing account. We also point out danger naive reasoning, which can lead to mis-estimation effects. illustrate these ideas graphical user interface have modeling.